避免在R data.table中使用sapply

Question

I've written a function to remove everything from the first parentheses onwards in a string:

until_parentheses <- function(string) {

  one <- stringr::str_split_fixed(string, "\\(", 2)[1, 1]

  res <- stringr::str_trim(one)

  return(res)

}

And I have a data.table with a column that looks (something) like this:

messy <- paste(letters[1:10], paste0(c(" (", letters[1:2], ")"), collapse = ""))

dt <- data.table(messy)

When I try to use until_parentheses() on the messy column like so

dt[, ":=" (clean = until_parentheses(messy))]

The function is applied to only the first element of messy and the clean column is the result repeated 10 times.

In order to have the clean column come out how I want it to I am using sapply:

dt[, ":=" (clean_2 = sapply(messy, until_parentheses))]

This gives the result I want however it takes a long time to run when dt is long.

I feel like there are problems with both my until_parenthese() function and with my data.table method. Does anyone have a solution that makes redundant my use of sapply in this instance?

Thanks!

Answer 1

您可以使用向量化的gsub ：

dt[,clean_3:=gsub(' +[(].*','',messy)] ## replace anything after the first ( with a blank